From: Greg Kroah-Hartman Date: Mon, 9 May 2022 14:36:35 +0000 (+0200) Subject: 5.17-stable patches X-Git-Tag: v4.9.313~51 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=fd688bc7e96aba6f39a1d54805ca3346d78ccc19;p=thirdparty%2Fkernel%2Fstable-queue.git 5.17-stable patches added patches: btrfs-always-log-symlinks-in-full-mode.patch --- diff --git a/queue-5.17/btrfs-always-log-symlinks-in-full-mode.patch b/queue-5.17/btrfs-always-log-symlinks-in-full-mode.patch new file mode 100644 index 00000000000..0fc37d1aabf --- /dev/null +++ b/queue-5.17/btrfs-always-log-symlinks-in-full-mode.patch @@ -0,0 +1,89 @@ +From d0e64a981fd841cb0f28fcd6afcac55e6f1e6994 Mon Sep 17 00:00:00 2001 +From: Filipe Manana +Date: Thu, 21 Apr 2022 10:56:39 +0100 +Subject: btrfs: always log symlinks in full mode + +From: Filipe Manana + +commit d0e64a981fd841cb0f28fcd6afcac55e6f1e6994 upstream. + +On Linux, empty symlinks are invalid, and attempting to create one with +the system call symlink(2) results in an -ENOENT error and this is +explicitly documented in the man page. + +If we rename a symlink that was created in the current transaction and its +parent directory was logged before, we actually end up logging the symlink +without logging its content, which is stored in an inline extent. That +means that after a power failure we can end up with an empty symlink, +having no content and an i_size of 0 bytes. + +It can be easily reproduced like this: + + $ mkfs.btrfs -f /dev/sdc + $ mount /dev/sdc /mnt + + $ mkdir /mnt/testdir + $ sync + + # Create a file inside the directory and fsync the directory. + $ touch /mnt/testdir/foo + $ xfs_io -c "fsync" /mnt/testdir + + # Create a symlink inside the directory and then rename the symlink. + $ ln -s /mnt/testdir/foo /mnt/testdir/bar + $ mv /mnt/testdir/bar /mnt/testdir/baz + + # Now fsync again the directory, this persist the log tree. + $ xfs_io -c "fsync" /mnt/testdir + + + + $ mount /dev/sdc /mnt + $ stat -c %s /mnt/testdir/baz + 0 + $ readlink /mnt/testdir/baz + $ + +Fix this by always logging symlinks in full mode (LOG_INODE_ALL), so that +their content is also logged. + +A test case for fstests will follow. + +CC: stable@vger.kernel.org # 4.9+ +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Greg Kroah-Hartman +--- + fs/btrfs/tree-log.c | 14 +++++++++++++- + 1 file changed, 13 insertions(+), 1 deletion(-) + +--- a/fs/btrfs/tree-log.c ++++ b/fs/btrfs/tree-log.c +@@ -5656,6 +5656,18 @@ static int btrfs_log_inode(struct btrfs_ + } + + /* ++ * For symlinks, we must always log their content, which is stored in an ++ * inline extent, otherwise we could end up with an empty symlink after ++ * log replay, which is invalid on linux (symlink(2) returns -ENOENT if ++ * one attempts to create an empty symlink). ++ * We don't need to worry about flushing delalloc, because when we create ++ * the inline extent when the symlink is created (we never have delalloc ++ * for symlinks). ++ */ ++ if (S_ISLNK(inode->vfs_inode.i_mode)) ++ inode_only = LOG_INODE_ALL; ++ ++ /* + * This is for cases where logging a directory could result in losing a + * a file after replaying the log. For example, if we move a file from a + * directory A to a directory B, then fsync directory A, we have no way +@@ -6015,7 +6027,7 @@ process_leaf: + } + + ctx->log_new_dentries = false; +- if (type == BTRFS_FT_DIR || type == BTRFS_FT_SYMLINK) ++ if (type == BTRFS_FT_DIR) + log_mode = LOG_INODE_ALL; + ret = btrfs_log_inode(trans, BTRFS_I(di_inode), + log_mode, ctx); diff --git a/queue-5.17/series b/queue-5.17/series index 8bbf482ee0e..0369ce6cd50 100644 --- a/queue-5.17/series +++ b/queue-5.17/series @@ -101,3 +101,4 @@ bnxt_en-fix-possible-bnxt_open-failure-caused-by-wrong-rfs-flag.patch bnxt_en-fix-unnecessary-dropping-of-rx-packets.patch selftests-ocelot-tc_flower_chains-specify-conform-exceed-action-for-policer.patch smsc911x-allow-using-irq0.patch +btrfs-always-log-symlinks-in-full-mode.patch